计算机应用 ›› 2014, Vol. 34 ›› Issue (10): 2944-2947.DOI: 10.11772/j.issn.1001-9081.2014.10.2944

• 虚拟现实与数字媒体 • 上一篇    下一篇

基于改进快速鲁棒特征的图像快速拼接算法

朱琳,王莹,刘淑云,赵博   

  1. 中国北方车辆研究所 信息与控制技术部,北京 100072
  • 收稿日期:2014-05-12 修回日期:2014-06-22 出版日期:2014-10-01 发布日期:2014-10-30
  • 通讯作者: 朱琳
  • 作者简介:朱琳(1990-),女,山东烟台人,硕士研究生,主要研究方向:机器学习、计算机视觉、图像拼接;王莹(1972-),女,湖北洪湖人,研究员,硕士,主要研究方向:计算机视觉、模式识别、目标探测与跟踪;刘淑云(1977-),女,河北唐山人,高级工程师,硕士,主要研究方向:图像处理、DSP算法实现;赵博(1984-),男,陕西延安人,工程师,博士,主要研究方向:计算机视觉、人工智能与机器学习、数字图像理解。
  • 基金资助:

    国家自然科学基金资助项目

Fast image stitching algorithm based on improved speeded up robust feature

ZHU Lin,WANG Ying,LIU Shuyun,ZHAO Bo   

  1. Department of Information and Control, China North Vehicle Research Institute, Beijing 100072, China
  • Received:2014-05-12 Revised:2014-06-22 Online:2014-10-01 Published:2014-10-30
  • Contact: ZHU Lin

摘要:

针对快速鲁棒特性(SURF)算法实时性、鲁棒性等无法满足实际应用需求的问题,提出了一种对SURF的改进算法,实现图像快速拼接。改进的算法采用机器学习的方法,建立一个二进制分类器,识别出SURF提取的特征点中的关键特征点,并剔除非关键特征点。此外,采用Relief-F算法将改进的SURF描述子降维简化来完成图像配准。图像融合阶段采用带阈值的加权融合算法,实现了图像无缝拼接。实验结果表明,改进的算法具有较强的实时性和鲁棒性,并且提高了图像配准的效率,加快了图像拼接的速度。

Abstract:

An fast image stitching algorithm based on improved Speeded Up Robust Feature (SURF) was proposed to overcome the real-time and robustness problems of the original SURF based stitching algorithms. The machine learning method was adopted to build a binary classifier, which identified the critical feature points obtained by SURF and removed the non-critical feature points. In addition, the Relief-F algorithm was used to reduce the dimension of the improved SURF descriptor to accomplish image registration. The weighted threshold fusion algorithm was adopted to achieve seamless image stitching. Several experiments were conducted to verify the real-time performance and robustness of the improved algorithm. Furthermore, the efficiency of image registration and the speed of image stitching were improved.

中图分类号: